Acoustic test chambers have been known in the art for many years. For example, in U.S. Pat. No. 3,014,543 an Acoustical Vibration Test Device was disclosed which subjected devices under test to high sound pressure levels, and was specifically aimed at testing aircraft and missile components. This patent disclosed the fact that random noise could be used by varying the spectral distribution of the sounds produced. One limitation of the apparatus disclosed in that patent was the use of two chambers (an outer and an inner chamber), one (the inner) to contain the device under test and to confine the sound generated by a loudspeaker, and the other (the outer) to produce low frequency sound.
In another U.S. Pat. No. 3,198,007, an acoustic testing chamber was disclosed which subjected devices under test to high intensity acoustic vibrations, and was specifically aimed at testing guided missile or rocket components. This patent disclosed the use of two sound sources, one located on the longitudinal axis of the test chamber, and the other located on an axis 90.degree. from the longitudinal axis. Also disclosed was the use of different frequency ranges produced by each of the two sound sources, and the use of an acoustical mirror to mix the sounds produced by the two sound sources and then direct such mixed sound onto the device under test.
In the electronics industry, a customary step of the manufacturing process is to assemble a printed circuit board with electrical and electronic components. Once the board is assembled, it is in a condition which lends itself to be tested, both mechanically and electrically. Typical mechanical tests that are performed on printed circuit boards at this stage of manufacture are sine wave vibration testing (typically at 2.2 g's at a non-resonant frequency or at a "swept" frequency) and temperature cycle testing (typically using thermal cycles from -40.degree. F. to +160.degree. F.). Such vibration testing does not simulate field operating conditions, nor does it act as an effective screening procedure by testing components under a wide range of non-resonant frequencies and g-forces.
The use of random vibration has been proven to be much more effective than the use of sine vibration. This is due to the presence of the entire frequency range. With sine vibration only one mode of vibration is excited at a time and only for a short period of time as the frequency is being swept. But with random vibration all the modes are excited. The premise is that random vibration screening is a cost saving both to the customer, in lower cost of ownership, and to the producer, in lower production costs. The key to the random vibration is the stimulation of the assembly maximizing the number of modes which can be excited during exposure to vibration. A final assembly which is well designed will minimize the number of modes and limit the displacement of the printed circuit assembly during exposure to vibration.